The Pharmacology of Continuous Infusions: Kinetics in Critical Illness

A Review for Critical Care Postgraduates

Dr Neeraj Manikath , clude.ai

Abstract

Continuous drug infusions form the cornerstone of hemodynamic and sedation management in critical illness. However, the profound pathophysiological derangements in critically ill patients fundamentally alter both pharmacokinetics (PK) and pharmacodynamics (PD), making standard dosing paradigms inadequate and potentially dangerous. This review explores the mechanisms underlying altered drug behavior in critical illness, with practical guidance for optimizing continuous infusions of sedatives, analgesics, and vasoactive medications in patients with organ dysfunction.

Introduction

The intensive care unit (ICU) represents a unique pharmacological challenge. Unlike stable outpatients, critically ill patients exhibit dramatic alterations in drug absorption, distribution, metabolism, and elimination. These changes are compounded by dynamic pathophysiology—what applies at admission may be irrelevant 48 hours later. For continuous infusions, understanding these principles is paramount, as steady-state concentrations depend on clearance, and therapeutic effect depends on both drug concentration at the effect site and receptor responsiveness.

The consequences of pharmacological ignorance in the ICU are significant: undersedation leads to patient-ventilator dyssynchrony and accidental extubation; oversedation prolongs mechanical ventilation; inadequate vasoactive support results in organ hypoperfusion; and excessive vasopressor use causes digital ischemia and bowel necrosis. This review provides a mechanistic framework for rational drug dosing in critical illness.

Altered Pharmacokinetics: The Foundation of Dosing Errors

Volume of Distribution: The "Moving Target"

Volume of distribution (Vd) represents the theoretical volume into which a drug distributes to produce the observed plasma concentration. In critical illness, Vd can change dramatically, particularly for hydrophilic drugs.

Capillary Leak Syndrome

Sepsis, severe trauma, burns, and major surgery trigger systemic inflammation with endothelial glycocalyx degradation and increased capillary permeability. The resulting third-spacing dramatically increases Vd for hydrophilic drugs such as beta-lactam antibiotics, aminoglycosides, and vancomycin. Studies demonstrate that Vd for aminoglycosides can increase from 0.25 L/kg in healthy individuals to >0.4 L/kg in septic shock, necessitating higher loading doses to achieve therapeutic concentrations.¹

Pearl: For hydrophilic drugs requiring rapid effect (e.g., antimicrobials), consider increasing the loading dose by 25-50% in patients with capillary leak, but maintain standard maintenance infusions as clearance may not proportionally increase.

Hypoalbuminemia: The Great Liberator

Albumin normally binds 90-95% of propofol, 85-90% of midazolam, and >99% of highly protein-bound drugs like ertapenem. In critical illness, albumin levels commonly fall below 25 g/L. This increases free (pharmacologically active) drug fraction, paradoxically requiring lower total drug concentrations to achieve effect.²

For example, a patient with albumin of 15 g/L receiving propofol will have approximately twice the free drug concentration compared to a patient with normal albumin receiving the same dose. This explains the increased sedation sensitivity and prolonged emergence often seen in hypoalbuminemic patients.

Oyster: Don't reflexively increase sedative infusions in hypoalbuminemic patients who appear "oversedated"—they may have therapeutic free drug levels despite subtherapeutic total levels. Consider reducing doses and reassessing clinically.

Fluid Resuscitation and Dynamic Changes

Aggressive fluid resuscitation—the median ICU patient receives 4-8 liters in the first 24 hours—profoundly increases Vd during the acute phase. However, as capillary integrity restores and diuresis occurs, Vd may normalize or even decrease below baseline. This dynamic nature means that a drug dose appropriate on day 1 may be excessive on day 3.³

Clearance: The Rate-Limiting Step

For continuous infusions at steady state, the infusion rate equals the elimination rate. Clearance (CL) is therefore the primary determinant of the maintenance dose required.

Renal Clearance: Beyond Creatinine

Serum creatinine is a notoriously poor marker of renal function in the ICU. Reduced muscle mass, increased volume of distribution, and delayed steady-state achievement mean that "normal" creatinine may mask significant renal impairment. Conversely, augmented renal clearance (ARC)—defined as creatinine clearance >130 mL/min/1.73m²—occurs in 30-65% of ICU patients, particularly younger trauma patients, leading to subtherapeutic levels of renally eliminated drugs.⁴

For continuous venovenous hemofiltration (CVVH) or hemodialysis, drug removal depends on molecular weight, protein binding, and dialysis parameters. Small, hydrophilic, minimally protein-bound drugs (e.g., morphine-6-glucuronide, active metabolite of morphine) are readily cleared, while highly protein-bound drugs (e.g., propofol) are minimally affected.

Hack: For patients on CVVH, assume approximately 30-40 mL/min of additional "renal" clearance for hydrophilic, low-protein-binding drugs. Increase maintenance infusions by 30% as a starting point and titrate clinically.

Hepatic Metabolism: The Multi-Organ Failure Multiplier

The liver eliminates drugs via two mechanisms:

Flow-dependent drugs (e.g., propofol, fentanyl, morphine): Clearance limited by hepatic blood flow
Capacity-dependent drugs (e.g., midazolam, lorazepam): Clearance limited by enzyme activity

In septic shock with reduced cardiac output and hepatosplanchnic hypoperfusion, clearance of flow-dependent drugs may decrease by 40-60%.⁵ Additionally, inflammatory cytokines (IL-6, TNF-α) downregulate CYP450 enzyme expression, reducing capacity-dependent clearance. The combination creates a "perfect storm" for drug accumulation.

Propofol infusion syndrome—a potentially fatal complication characterized by metabolic acidosis, rhabdomyolysis, cardiac failure, and death—occurs more commonly with prolonged high-dose infusions (>4 mg/kg/h for >48 hours), particularly in patients with impaired clearance.⁶

Pearl: In shock states with vasopressor dependence, expect 30-50% reductions in hepatic clearance. Start sedative infusions at lower doses and titrate slowly, particularly with propofol. Consider switching from propofol to dexmedetomidine after 48 hours if high doses are required.

Altered Pharmacodynamics: When Receptors Stop Listening

While PK alterations explain why drug concentrations differ from expected, PD changes explain why patients respond differently to the same concentration.

Receptor Downregulation and Desensitization

Catecholamine Receptor Dysfunction in Sepsis

Septic shock is characterized by profound β-adrenergic receptor desensitization. Mechanisms include:

G-protein receptor kinase (GRK)-mediated phosphorylation and internalization
Uncoupling of receptors from downstream signaling
Reduced receptor density on cardiomyocyte membranes

This explains the clinical observation that septic patients often require escalating norepinephrine doses (frequently >0.5 mcg/kg/min) despite adequate resuscitation, a phenomenon rarely seen in other forms of distributive shock.⁷

Hack: When norepinephrine doses exceed 0.3-0.5 mcg/kg/min in septic shock, consider adding vasopressin (0.03-0.04 units/min) or angiotensin II. These agents work via non-adrenergic mechanisms and can restore MAP while allowing norepinephrine dose reduction.

Tolerance Development

Continuous benzodiazepine infusions induce rapid tolerance via GABA receptor downregulation and altered subunit composition. Propofol shows less tolerance, while dexmedetomidine, acting via α2-receptors, demonstrates minimal tolerance development even with prolonged infusions.⁸

Oyster: If sedation requirements are escalating rapidly (>20% increase per day), suspect tolerance rather than inadequate dosing. Consider rotating sedative classes (e.g., benzodiazepine to propofol or dexmedetomidine) rather than continually escalating a single agent.

The Inflammatory Milieu

Pro-inflammatory cytokines alter receptor expression and sensitivity. IL-1β and TNF-α reduce opioid receptor expression and increase production of anti-opioid peptides, contributing to the higher analgesic requirements observed in septic patients.⁹

Clinical Application: Practical Dosing Strategies

Sedatives in Organ Dysfunction

Propofol

Normal function: 25-75 mcg/kg/min
Liver dysfunction: Start 50% lower; hepatic metabolism impaired
Renal dysfunction: Minimal impact on parent drug, but monitor for propofol infusion syndrome (acidosis, triglycerides)
Titration tip: Increase by 5-10 mcg/kg/min increments every 10-15 minutes

Pearl: Check triglycerides daily with propofol >50 mcg/kg/min. Levels >400 mg/dL warrant dose reduction or agent change.

Midazolam

Normal function: 0.5-4 mg/h
Liver dysfunction: Reduce by 50%; active metabolites accumulate
Renal dysfunction: Active metabolite (1-hydroxy-midazolam) accumulates; avoid for prolonged infusions
Titration tip: Bolus 1-2 mg, then infusion; reassess every 2-4 hours

Hack: For difficult-to-sedate patients, combine low-dose midazolam (1-2 mg/h) with propofol rather than escalating a single agent. The synergistic effect often allows lower total doses of each.

Dexmedetomidine

Normal function: 0.2-1.5 mcg/kg/h
Liver dysfunction: Reduce by 25%; hepatic metabolism
Renal dysfunction: No dose adjustment needed
Unique advantage: Minimal respiratory depression; useful for non-intubated patients and liberation from mechanical ventilation

Oyster: Dexmedetomidine causes bradycardia and hypotension in 20-30% of patients. Avoid loading doses in hemodynamically unstable patients. Start at 0.2 mcg/kg/h and titrate slowly.

Analgesics in Organ Dysfunction

Fentanyl

Normal function: 25-200 mcg/h
Liver dysfunction: Reduce by 25-50%; flow-dependent clearance
Renal dysfunction: No active metabolites; preferred opioid in renal failure
Context-sensitive half-time: Increases with prolonged infusions (>48h); expect delayed awakening

Morphine

Renal dysfunction: AVOID prolonged infusions; morphine-6-glucuronide (M6G) accumulates, causing prolonged sedation and respiratory depression
Hack: If morphine is used in renal failure, reduce dose by 75% and consider intermittent boluses rather than continuous infusion

Remifentanil

Unique property: Metabolized by plasma esterases; independent of organ function
Dose: 0.05-0.2 mcg/kg/min
Advantage: Rapid offset (3-10 minutes) regardless of infusion duration
Oyster: Remifentanil is ideal for patients requiring frequent neurological assessments or anticipated extubation. However, establish alternative analgesia before discontinuation to prevent rebound pain.

Vasoactive Drugs in Organ Dysfunction

Norepinephrine

Normal starting dose: 0.05-0.1 mcg/kg/min
Organ dysfunction impact: Minimal—catecholamine metabolism occurs in multiple tissues
Titration: Increase by 0.05 mcg/kg/min every 3-5 minutes to MAP target (typically 65 mmHg)

Pearl: In profound shock requiring >0.5 mcg/kg/min norepinephrine, check lactate and ScvO2 to ensure adequate cardiac output. Peripheral vasoconstriction may normalize MAP while masking inadequate tissue perfusion.

Vasopressin

Dose: Fixed at 0.03-0.04 units/min (NOT titrated)
Renal dysfunction: No adjustment needed
Unique benefit: Restores vascular tone in catecholamine-resistant shock; beneficial in hepatorenal syndrome

Milrinone

Normal dose: 0.25-0.75 mcg/kg/min (after 50 mcg/kg loading dose)
Renal dysfunction: Reduce by 50-75% (primarily renal elimination)
Hack: In renal failure, skip the loading dose and start at 0.125 mcg/kg/min. Titrate over 6-12 hours rather than acutely.

Dobutamine

Hepatic dysfunction: Minimal impact
Renal dysfunction: No adjustment
Tolerance: Develops within 48-72 hours; consider intermittent dosing or rotation if prolonged support needed

Practical Integration: The "Daily Dose Check"

Implement a systematic daily assessment:

Volume Status: Fluid overloaded? Consider increased Vd—may need higher loading doses but standard maintenance
Albumin Level: <25 g/L? Expect increased free drug fraction for protein-bound drugs
Renal Function: Calculate estimated CrCl; adjust renally cleared drugs; watch for ARC in young patients
Liver Function: Elevated bilirubin, INR, or lactate? Reduce hepatically cleared drugs by 25-50%
Inflammatory State: High CRP, procalcitonin? Expect altered PD—may need higher vasoactive/analgesic doses
Dialysis Status: CVVH? Add 30-40 mL/min to estimated CrCl for hydrophilic drugs

Hack: Create a daily "drug clearance assessment" using this mnemonic: VALID = Volume, Albumin, Liver, Inflammation, Dialysis/renal function

Special Populations

Obesity

Use ideal body weight (IBW) for hydrophilic drugs and adjusted body weight (ABW = IBW + 0.4[TBW - IBW]) for lipophilic drugs. Propofol should be dosed on total body weight but not exceed 80 mcg/kg/min based on ideal weight to prevent toxicity.

Elderly

Increased Vd for lipophilic drugs due to increased body fat; decreased clearance for hepatically metabolized drugs; increased sensitivity to sedatives. Start doses 25-50% lower.

Burns

Dramatically increased clearance (up to 2-3x normal) for most drugs due to hypermetabolic state and increased hepatic blood flow. May require 50-100% dose increases.

Monitoring and Titration Strategies

Clinical Endpoints Over Kinetics

While therapeutic drug monitoring (TDM) is valuable for antibiotics, continuous infusion sedatives and vasoactive agents require clinical titration:

Sedation: Use validated scales (RASS, SAS); target light sedation (RASS -1 to 0) unless specific indications for deep sedation
Analgesia: Use behavioral pain scales (BPS, CPOT) in non-communicative patients
Hemodynamics: MAP is a surrogate; ensure adequate perfusion (lactate clearance, urine output, capillary refill)

Pearl: The best drug level is the one that achieves clinical effect without toxicity. Algorithms are starting points—individual titration is essential.

Conclusion

Continuous drug infusions in critical illness require abandoning the "cookbook" approach. Understanding the mechanistic basis of altered PK (increased Vd from capillary leak, reduced clearance from organ dysfunction, hypoalbuminemia effects) and PD (receptor downregulation, tolerance, inflammatory modulation) allows rational, individualized dosing.

Key principles:

Front-load for effect: Adequate loading doses accounting for increased Vd
Maintain with caution: Reduce maintenance doses in proportion to clearance reduction
Reassess daily: Organ function and volume status change dynamically
Monitor clinically: Titrate to effect, not to formula
Plan for tolerance: Rotate agents or use multimodal approaches for prolonged needs

The art of ICU pharmacology lies in recognizing that critically ill patients are not simply "very sick" versions of healthy people—they are pharmacologically distinct entities requiring an evidence-based but flexible approach to continuous infusions.

References

Udy AA, Roberts JA, Lipman J. Clinical implications of antibiotic pharmacokinetic principles in the critically ill. Intensive Care Med. 2013;39(12):2070-2082.
Roberts JA, Pea F, Lipman J. The clinical relevance of plasma protein binding changes. Clin Pharmacokinet. 2013;52(1):1-8.
Malbrain ML, Marik PE, Witters I, et al. Fluid overload, de-resuscitation, and outcomes in critically ill or injured patients: a systematic review with suggestions for clinical practice. Anaesthesiol Intensive Ther. 2014;46(5):361-380.
Udy AA, Roberts JA, Boots RJ, et al. Augmented renal clearance: implications for antibacterial dosing in the critically ill. Clin Pharmacokinet. 2010;49(1):1-16.
Novovic M, Aleksic A, Novovic E, et al. Pharmacokinetic changes in critical illness. Med Pregl. 2013;66(7-8):279-285.
Kam PC, Cardone D. Propofol infusion syndrome. Anaesthesia. 2007;62(7):690-701.
Dünser MW, Hasibeder WR. Sympathetic overstimulation during critical illness: adverse effects of adrenergic stress. J Intensive Care Med. 2009;24(5):293-316.
Devlin JW, Roberts RJ. Pharmacology of commonly used analgesics and sedatives in the ICU: benzodiazepines, propofol, and opioids. Crit Care Clin. 2009;25(3):431-449.
Beilin B, Bessler H, Mayburd E, et al. Effects of preemptive analgesia on pain and cytokine production in the postoperative period. Anesthesiology. 2003;98(1):151-155.

Disclosure: The author declares no conflicts of interest.

For Further Reading:

Society of Critical Care Medicine Clinical Practice Guidelines for Sustained Neuromuscular Blockade in the Adult Critically Ill Patient (2016)
Barr J, et al. Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med. 2013;41(1):263-306.
Vincent JL, et al. Circulatory shock. N Engl J Med. 2013;369(18):1726-1734.

learningmedicine

Wednesday, November 12, 2025